Theory of Cones

George Phillip
Department

Barker

ofMathemutics
ofMissouri-Kansas

University
Kcmsas City, Missouri 64110

City

Submittedby Hans Schneider

ABSTRACT
This survey deals with the aspects of archimedian
partially ordered finitedimensional real vector spaces and order preserving linear maps which do not involve
spectral theory. The first section sketches some of the background
of entrywise
nonnegative
matrices and of systems of inequalities which motivate much of the
current investigations.
The study of inequalities
resulted in the definition
of a
polyhedral cone K and its face lattice F(K). In Section 1I.A the face lattice of a not
necessarily polyhedral cone K in a vector space V is investigated.
In particular the
interplay between the lattice properties of O!(K) and geometric properties of K is
emphasized. Section 1I.B turns to the cones II(K) in the space of linear maps on V.
Recall that II(K) is the cone of all order preserving linear maps. Of particular interest
are the algebraic structure of II(K) as a semiring and the nature of the group Aut( K )
of nonsingular elements A E II( K) for which A E II( K) as well. In a short final
section the cone p,, of n X n positive semidefinite matrices is discussed. A characterization of the set of completely positive linear maps is stated. The proofs will appear in a
forthcoming paper.

I.

HISTORICAL

NOTES

In 1907 Perron [56] initiated the study of matrices A =(aii) all of whose
entries are positive, and Frobenius
[32, 331 carried this study further. If
p=p(A)=max(]Xj:
h is an eigenvalue of A}, then their results showed that p
is an eigenvalue of A which is a simple root of the characteristic
polynomial
and that the eigenvector
belonging to p may be taken to have all entries
positive. In 1912 Frobenius [34] showed that these results hold for a broader
class of matrices with nonnegative
entries. These are the irreducible matrices.

LlNEAR ALGEBRA AND ITS APPLICATIONS 39:263-291

(1981)

263

c Elsevier North Holland, Inc., 1981

52 Vande+ilt

Ave., New York, NY 10017

0024-3795,81/050263

+ 29$02.50

264

GEORGE PHILLIP BARKER

A matrix A is called reducible

iff there is a pemmtation matrix P such that

PTAP=

AI

A,

where A, and A, are square matrices.

The study of nonnegative matrices remained dormant until 1950. Of
course, certain special types such as stochastic matrices were investigated
during this period, but Wieldandts paper [75] seems to have attracted
attention to this area again.
Kantorovich in 1935 published an often overlooked paper [44] concerning
operators on a partially ordered Banach space. Krein and Rutman [45] picked
up this thread. At this point we require a definition.
DEFINITION 1.1. Let V be a real topological vector space. A cone K in V
is a subset of V which satisfies
(i) if a,P>O and x, ~EK,
(ii) Kn(-K)=(o).

then ox+byEK,

We shall also require that K be a closed set. The cone K is fulliff int K # 0,
and K is reproducing iff K-K = V.
REMARK 1.2. In the case of a finite-dimensional V (the situation with
which we shall be principally concerned) a cone K is full iff it is reproducing.
Krein and Rutman considered operators which leave invariant a closed
cone in a Banach space. With certain restrictions on the operators, Krein and
Rutman obtained extensions of the Perron-Frobenius theorems. Birlchoff 1221
gave a proof of the finite-dimensional case of this result. Vandergraft 1721 also
investigated the problem with an eye to the applications to the convergence
of certain iterative processes. These applications have been fruitful, and one
can consult Schrijder [62], Trottenberg and Winter [70], and Vandergraft [73]
for the developments along these lines. The spectral theory of cone preserving
operators is now fairly complete and has been summarized in Barker and
Schneider [14] and Berman and Plemmons [21, Chapter 11. Since the subject
of spectral theory is well summarized, we shall not consider it in detail in the
main body of this paper. However, in order to have a complete extension of
the Perron-Frobenius results it is necessary to have for cone preserving maps
an analog of irreducibility. This extension is based on certain distinguished
subsets of K, namely its faces. The following definition is due to Hans
Schneider.

THEORY OF CONES

265

DEFINITION 1.3. Let K be a closed cone in the finite-dimensional

space V. A cone F C K is a fuce of K iff

real

XEK,

y--x~K,and

~EF

imply

XEF.

The collection of all faces of K is denoted by $7(K) or simply :y.

The trivial faces are {0} and K. If FE?wewriteFdK,andif
and FCG, then FdG (cf. [4]).

FaK,

GaK,

DEFINITION 1.4. A linear map A which leaves K invariant is irre&cihZe

iff A leaves invariant no nontrivial face of K.
The study of linear inequalities has led to the study of a special class of
cones, the polyhedral cones (cf. Definition 1.5). This study, which dates at
least from Farkass paper of 1902 (see chapters 1 and 2 of Stoer and Witzgall
[63] and their bibliography), has been applied to econometrics and mathematical programming. Before discussing this we need some notation.
Let K be a cone in V. If XEK, we write ~30. Then x3 y means x- ~20.
Also x>O and x>O mean x#O, XE K, and x~int K respectively. Thus K
defines a partial order on V. Let K be a closed full cone, and let V and Horn
(V) be the dual space of V and the space of linear transformation on V
respectively.
Set
K={f~V:fiz0Vx~K},
II(K)={AE

Horn(V):

AKCK}.

Then K and II(K) are closed full cones in V and Horn (V)
when K is a closed full cone. Thus f 30 means fE K, and
A E I1( K ). The dimension of a face F, dim F, is the dimension
span F-F of F. If SC K, then Q(S) will denote the least face of
S, viz.,
Q(S)=

f-l {F: ScF,

respectively
A >O means
of the linear
K containing

FaK}.

If S= {x), we write Q(x). If dim F= 1, F is called an extreme ray of K, and if

x30 is such that dim@(x)= 1, then x is called an extremal. One can show
that if x is an extremal and 0 < y < x, then there is an (Y> 0 such that cuy= X.
Also one can show that K is the convex hull of its extreme rays. (See Barker
[2, 31 and Glazman and Ljubie [36]).

266

GEORGE PHILLIP BARKER

Let f be a linear functional on Iw, and consider the homogeneous

inequality f( x)20.
This defines a half space in Iw, and a finite set of
inequalities is satisfied by any point in the intersection of the corresponding
half spaces. Such an intersection need not be a cone, for it need not be
pointed, nor is such an intersection necessarily full. However, we shall be
interested only in those which are.
DEFINITION 1.5.
extreme rays.

A polyhedral

cone is a cone which has a finite set of

The literature on polyhedral cones is extensive.

begin with the following references: Gerstenhaber
[37], Stoer and Witzgall [63], and Uzawa [71]. We
the basic results for polyhedral cones (Uzawa [71,
THEOREM 1.6 (Representation
the intersection

theorem).

The interested reader can

[35], Goldman and Tucker
state without proof two of
p. 281).

A cone K is polyhedral

iff it is

of a finite set of half spaces.

THEOREM 1.7 (Duality theorem).

For any polyhedral
cone K is also polyhedral, and K= K.

cone K, the dual

These results answer questions which arise naturally in studying linear

inequalities. However, when considering cone preserving maps it is reasonable to inquire about the relation of K to II{ K). We shall investigate this in
some detail in Section II.B, but for the moment let us note one result.
THEOREM 1.8 (Schneider and Vidyasagar [61], Tam [67]).
polyhedral

The cone K is

iff II( K) is polyhedral.

Not surprisingly, the theory of polyhedral cones is intimately related to the

theory of convex polyhedra. Given a cone K and a hyperplane H= {x: fx= 1
for some fixed f E V} which meets the interior of K, the intersection Hn K is
a cross section of K which is a convex polyhedron. This process can be
reversed to obtain a cone in (w from a convex body in I%- (cf. Stoer and
Witzgall [63, pp. 73-741). Inspection shows that this process does not depend
upon the cones being polyhedral. Thus for any cone K the set %( K ) becomes
a lattice if for any F, GET we set

and
FvG=@(FUG).

267

THEORY OF CONES

The face lattice of a convex polyhedron is well known (cf. Griinbaum [38] or
McMullen and Shephard [48]). Using the process previously described, we
may translate a theorem about the face lattice of a convex polyhedron into a
theorem about F(K) for a polyhedral cone K and vice versa. We shall say a
great deal more about the lattice F(K) in the next section. For the present we
note one result.
THEOREM 1.9 (Barker [4], Stoer and Witzgall [63]).
5(K)

is a complemented

lattice. Zf K is polyhedral,

Zf K is a cone, then
then 4(K)
is rekztively

complemented,

If K is not polyhedral, then 5(K) is not in general relatively complemented. The reader may refer to Birkhoff [23] for various terms from lattice
theory.

II.

STRUCTURE

THEORY

This section consists of two parts. In the first we consider the structure of
and in the second that of II(K).
However, when it seems
appropriate to do so we shall consider the relationships of K and II(K) in
subsection A. Throughout this section we assume that K is a full closed
pointed cone in V.
K and K,

A.

The Theory for K

As has been noted in Sec. I, T(K)

contains the basic arithmetic of faces.
PROPOSITION2.A.l
nonempty

is a lattice. The next proposition

(Barker [4], Barker and Carlson [lo]).

Let S, T be

subsets of K and x1,. . . , x, vectors in K. Then

We started with the cone WY (whlc h we call the nonnegative orthant) of

all vectors [x,,..., x,]~ for which xi 30. This cone is not only polyhedral, but
it has n extreme rays, where n is the dimension of the ambient vector space.
Since in general a polyhedral cone K C V will have k> n=dimV extreme rays,
we single out this special feature of r~:.
DEFINITION 2.A.2. If K is a cone in an n-dimensional vector space, we
call K simplicial iff K has exactly n extreme rays.

GEORGE PHILLIP BARKER

268

If K is simphcial, and if we take a representation of V as II%

n, then K is the
image of R+ under a nonsingular linear transformation. It is easily checked
that S(lR+) and hence 9(K) is distributive. This is the easy direction of the
next result.
THEOREM 2.A.3 (Barker [4]).

4(K)

is distributive

iff K is simplicial.

The proof utilizes a technique due to Hans Schneider.

Proof. For n = 1,2, all cones K are simplicial and 9?(K) is distributive.
Thus let n 33 and suppose %(K) is distributive. We show that there are n
distinct extreme rays. Call two extremals ur, ~a distinct if ia( oi)#@(uz).
Suppose to the contrary that K has at least n+ 1 distinct extremals
n=dim K, this set is linearly dependent, so we have a

A,x,S-

But K is pointed, so Xi >0

terms and let

...+h+lX+l=O.

for some i and Xi < 0 for some i. Drop the zero

Zi =
i

hiXi

if

Xi>O,

-Xixi

if

h,<O.

Then the zk are distinct extremals, and renumbering if necessary, we have

Zi + . . . fz,

=Zq+l+

. . . +z,.

By Proposition 2.A.l (ii) we have

If zi, zi are distinct extremals, then @(zi)r\Q( xi)=O. Using the distributivity
of %( K ), we have

THEORYOFCONES

269

This contradiction implies that there can be at most n extremals. However, K

n
is full so there must be at least n extremals. Thus K is simplicial.
A leitmotif has been conditions on ?7(K ), II( K ), or some other set related
to K which are equivalent to K simplicial. We shall see a number of examples
of this.
Perhaps the next strongest requirement to impose on Y!(K) is that of
modularity. That is, if F, G, HE??(K) and if FaH, then

Distributive lattices are of course modular. For a simple example of a modular

but not distributive face lattice, consider the following example. Let dimV>3,
and let
K=(xEV:x++;+

.--+E)12].

For n=4 this is the Lorentz cone of causal vectors of special relativity. It is
easily seen that every nonzero x E a K is an extremal and consequently ?( K )
is moddar. We can isolate this feature.
DEFINITION2.A.4. Let K be a cone in V. K is strictly convex iff for any
nonzero noncollinear x1, xs ~i3 K we have Q(x,)v@(x,)=K.
It is easily checked that for any strictly convex K, %( K ) is modular.
shall see later, the cone m X m of positive semidefinite matrices is a full
pointed cone in the real space of m Xm hermitian matrices and it
modular face lattice. But for m>2 this cone is not strictly convex.
n=dimV=3
we have the following converse.
THEOREM2.A.5 (Barker [4]).
not distributive,

Problem.

Zf dim K=3

As we
closed
has a
When

and if ?T(K) is modular

but

then K is strictly convex.

Classify those cones whose face lattices are modular.

The genera1 problem is in some sense still open, although we can deal with
some special cases. In another sense the problem is close to finished, as we
shall see later. It is reasonable to consider the polyhedral case first, and in this
situation we have a strong result. Note that for any K the lattice g(K) is of
finite length, so we may take as the definition of semimodularity Corollary 1
of Birkhoff [23, p. 811.

270

GEORGE PHILLIP BARKER

DEFINITION 2.A.6. A lattice L of finite length is semimodular

elements a, b EL, a covers a A b only if a v b covers b.
THEOREM2.A.7 (Barker [4]).
lar, then T(K)

If K is polyhedral

and F(K)

iff for all

is semimodu-

is distributive.

Proof. We shall establish that K is simplicial. If HU K, then %( H j =

(Fcg( K j : FdH}.
Clearly %( H j is a sublattice of T( K j and is semimodular

when VT(K) is semimodular. We induct on dim K. For dim K =2, K is

simplicial. Assume the result holds for all polyhedral cones of dimension tn,
x,_ 1 be any n - 1 distinct extremals of K. Then
andletdimK=n.Letr,,...,
@(x,)v . . . v@(x,_,)#
K. For if equality were to hold, then (by renumbering the xi) we could find an m<n-1
such that H=@(x,)v
. . .v@(x,_,)
is
a complement in the lattice T( K j to Q(x, j. But dim H< n, and T(H) is
semimodular, so by the induction hypothesis g(H) is distributive, that is, H is
But @(x,,)
the convex hull of the rays a( X, j, . . . , Ca(x,_,) and dim H=m-1.
covers @(x,,)AH=
(O}, so @(x,)vH
covers H. However, K is polyhedral, so
dim(Q(x,)vH)=
l+dim H=mfn1, while dim K=n.
This contradicts
Therefore F is simpli@(x,)vH=K.
Thus F=Q(x,)v
. . . vQ(x,_,)#K.
cial. Consequently the cone K is n - 1 neighborly and must itself be simplicial
(cf Griinbaum [38, pp. 122- 1231).
n
When K is polyhedral it is well known that T(K) and %( K) are
anti-isomorphic (McMullen and Shephard [48]). However, when K is not
polyhedral the geometry is more complicated. To explore this matter we wish
to associate with each face of K a face of K. For any set SC V let

SI
DEFINITION2.A.8.

={fEV:

fs=OVsES}.

If Fd K, we define the dual F D of F to be

F={f~k:

fx=OVxEF}.

Note that F D = K n F I. If G E ??( K ), we shall use the notation GS to denote

the dual of G in g(K).
Barker [73 and Tam [69] independently introduced the duality operators.
In an unpublished summary entitled On the structure of the cone of positive
operators Tam states without proof a number of results not only for the
operators between K and K but also for the operators between II(K) and
lI( K ). Some proofs can be found in Tams paper [69] previously mentioned,
and others are in his doctoral thesis submitted to the University of Hong
Kong.

THEORY OF CONES

271

For FdK where K is polyhedral we have F Ds= F, but for general cones K
this equality does not hold. The next definition singles out this property.
Let K be a cone and FQ K. F is exposed iff F D6= F.

DEFINITION2.A.9.

A face F is exposed, as one may easily verify, iff F is the intersection of a

hyperplane with K. That is, there is some f EK such that F = ker f n K. We
always have F D = F 9 and the operation g F= F Ds is a closure operation.
Thus the correspondences F-+ FD and G-+G define a Galois connection
between CY(K ) and 9( K). This is a generalization of the earlier remark that
for polyhedral K, T(K) and 9( K) are anti-isomorphic.
The duality operator is reasonably well behaved with respect to the lattice
operations.
LEMMA2.A.10.

lfF,, F2a K, then

(i) (F,vF,)~=F~AF$,
(ii) (F, AF,)~Q FfvFzD.
The proof is not difficult and can be found in Barker [7], where examples
are given to show that equality need not hold in (ii). Not surprisingly, equality
in property (ii) relates to whether faces are exposed. In particular we have the
following result.
THEOREM2.A. 11.

Every face in both 3(K) and T( K) is exposed iff both

(i) (F,AF,)~=F~vF~,
(ii) (G,~G,)s=G~vG~,

F,, F, ET(K),
G,,G,eT(K).

and

As with the polyhedral cones, every face of the Lorentz cone is exposed.
For the Lorentz cone K in R, if we represent the dual space by the same R,
where linear functionals act via the usual inner product, then K and K are
comparable. In fact, K =K. More generally, if V is an inner-product space,
we may identify V with V where f;c= (f, x ) . Whenever we consider R , we
shall assume the usual inner product so that the cones K and K lie in the
same ambient space.
DEFINITION2.A.12.
selfdual iff K = K.

Let V be an inner-product space. The cone K C V is

PROPOSITION2.A.13.
Let K be selfdual.
semimodular and every face of K is exposed.

Then g(K)

is modular iff it is

272

GEORGE PHILLIP BARKER

This proposition is rather weak, but there are examples of selfdual cones
for which G?(K) is semimodular but not modular and of self-dual polyhedral
cones which are not simplicial. The interesting part of the proof is the
following fact.
LEMMA 2.A.14.

Let K be selfdual.

If FdK,

then FD is a complement

ofF.

In the remainder of subsection A, with a few exceptions, we are concerned only with selfdual K. Consequently, the two duality operators coincide, and we use the notation FD for the dual face. Further, if FdK, then we
let F denote the cone dual of Fin span F. That is,

It is useful to recall that

spanF=F-F={fi-fi:f;EF,i=I,2}.
The next definition is due to Raphael Loewy, in a private communication.
DEFINITION2.A.15.

Let K be a self-dual cone. K is perfect iff VF ~3( K ),

F =F.

THEOREM 2.A.16 (Barker [B]). Let K be a selfdual cone. Then K is

iff 3( K ) is orthomodular under the duality operation F --+F D.

perfect

If K is perfect, a routine argument shows that 4(K) is orthomodular. For

the proof of the converse we need a technical lemma from Barker and
Schneider [14], whose proof is omitted.
LEMMA 2.A.17.
Let K be a cone. Let u>O,
and let -v@K.
Put
~=sup{(v: u--olv>O}. Then O<E<CO and u-EvE~K.
For the proof of the theorem let K be a self-dual cone whose face lattice
%(K) is orthomodular. Let FEY(K)
with F=@(x),
x>O. Clearly, FcF.
Suppose there is a y E F \ F. Then by Lemma 2.A. 17 there is an E>O such
that x + EY is in the relative boundary of F. If E= (1 - p )/j3 for fl E (O,l), then
z=pr+(l-/3)y=fi(x+ey)
is also in the relative boundary of F. Since z#O,
then @( =;)a
F. By assumption

THEORY OF CONES

whence

273

there exists a uE@(z)~AF,

O=(z,

u) =/3(x,

Thus (u, Z) ~0, SO

u#O.

24) +(l-_p)(Y,

u>>P(X> u) >O>

a contradiction,
Hence F = F, and K is perfect.
So far we have not considered the fine structure of cones; specifically,
whether there is any construction
analogous to the direct-sum vector spaces.
Haynsworth [39] considered an external direct sum, but Loewy and Schneider
[46] formulated
the extremely useful notion of decomposability,
which is
analogous to an internal direct sum. Here the cone K need not be self-dual.
DEFINITION 2.A.18.
Let K be a cone in the vector space V. Let K, and
K, be subsets of K. We say K is the direct sum of K, and I(, (and write
K=K,CBK,)

iff

(i) span K, nspan

(ii) K=K,+K
2

K, = {0},

The cone K is called decomposable if there exist nonzero

such that K = K,CB K,, and indecomposable otherwise.
In Lemma
that if K=K,@

3.2 of the aforementioned

K,, then K,, K, ET(K).

subsets K, and K,

paper, Loewy and Schneider

However, more can be said.

LEMMA 2.A.19.
Let K==K,@K,.
If XEK,,
~EK,
implies
then (K,@K,)=
K,G3K,. Zf K is selfdual,
then K, = K,.

(x,y)

show

=O,

The proof of the first assertion can be found in Berman [ 191 or Barker and
Foran [12]. The second assertion is discussed in Barker [8]. This lemma is
used in the characterization
of polyhedral perfect cones by Barker and Foran
P21.
THEOREM 2.A.20.
proper maximal face
nonnegative

Zf K is a selfdual
polyhedral cone such that every
is selfdual
in its span, then K is the image of the

orthant under an orthogonal

transformation.

Summary of the proof.

A self-dual simplicial cone is the image of the
nonnegative
orthant under an orthogonal transformation,
whence it is enough
to show that K is simplicial. First establish by induction that K is perfect, that
is, that every face of K is selfdual in its span. This is the longer portion of the
proof. Now let us show again by induction that K is simplicial. Suppose that
whenever dim K < n- 1 (and K is polyhedral) we have K perfect implies K

274

GEORGE PHILUP BARKER

simplicial. Let dim K=n. Let x be an extremal of K, and put F=@(x)~. Then
F is a simplicial cone of dimension n- 1. We claim that K=F@@(
x). Let
K,=F@Q(x).
By Lemma 2.A.19

Thus K,cK

and K,=K,zK=K.

Hence K=K,

and the theorem is proved.

n
Theorem (5.3) of Barker [7] is now a corollary of Theorems 2.A.16 and
2.A.20.
COROLLARY2.A.21.
orthomodular

Let K be a selfdual

polyhedral

cone. Then T(K)

is

iff K is simplicial.

It should be noted that there are self-dual polyhedral cones which are not
simplicial in any V with dimV>2. Examples and a method of construction
are given in [ 121.
Another corollary of Theorem 2.A.16, this time together with Proposition
2.A.13, is the following result.
COROLLARY2.A.22.

lf K is self-dual

and 4(K)

is modular,

then K is

perfect.

The converse is appealing, since it would classify the self-dual cones with
modular face lattices. The converse is true for n= 1,2,3, but false for n=4.
An example can be found in [9]. To classify modular face lattices, even for
self-dual cones K, we would like to reduce the problem. Since sublattices of
modular lattices are again modular, the next result proves useful.
THEOREM2.A.23 (Barker [B]).
Zf K=K,@K,,
there are faces

Let K be a full closed pointed cone in V.

then ~(K)=~(K,)G3~(K,).
K,, K, such that $ =OI(Ki),

Conversely, if %(K)=%I@??z,
i= 1,2, and K=K,@K,.

Proof.
The proof of the first statement is a routine verification, so let us
consider the second. If T(K)=91@??S,
then every extremal belongs to either
YTior $,. Let Ki be the join of all extreme rays in i. Since 3(K) is atomic, SO
is Fi, and thus $ = 9(Ki). Similarly K, +K, = K. To finish we must show that
span K, fl span K, = { 0). Obviously, K, n K, = { 0). Suppose z E span K, f?
span K,. For any wi EK,, ui EKE, i= 1,2, for which
.z=wl-w2=u1-u2,

THEORY OF CONES

275

we have

By the uniqueness of the representations in ?F=~r@$,

~(wr)=ip(w,)

and

it follows that

@(ui)=(a(ua).

(*>

So if
If z#O, then -ZE K,; otherwise -zEK,nspanK,=KznK,={O).
we take ui in the relative interior of K,, there is by Lemma 2.A.17 an E>O
such that ui - EZ= u2 is in the relative boundary of K,. Then

and by (*) a( or)=@.

This is impossible unless dimQ( u,)=dim K, = 1.
In this case -Z E K, and we have a contradiction. Hence z=O and K = K,G3 K,.
This result shows that a classification of indecomposable cones with
modular face lattices solves the problem. Let K be a cone whose face lattice
T(K) is modular. It is well known that every maximal chain (from {O} to K)
has the same length (Birkhoff [23]). This length, which is one less than the
number of faces in the chain, is frequently called the dimension of the lattice.
We use the term height for this length, and reserve dimension for the
algebraic dimension of the vector space spanned by a face. Thus the Lorentz
cone (of any dimension >l) has height 2. A useful property of lattices is
subdirect irreducibility (cf. Crawley and Dilworth [27]). Its relevance here is
expressed in the next lemma, from [9].
LEMMA 2.A.24. Suppose %(K)
iff K is indecomposable.

is modular.

Then

9(K)

is subdirectly

irreducible

This point is important because Crawley and Dilworth [27] present

coordinatization theorems (Theorems 13.4 and 13.5) which imply that if K is
an indecomposable cone for which T(K) is a modular lattice of height 34,
then Y(K) is isomorphic qua lattice to the lattice of subspaces of a vector
space over a division ring. The example of the cone of positive semidefinite
matrices in the real space of nXn hermitian matrices over the quaternions
shows that the division ring need not be a field even in our restricted setting.
The problem of classifying the modular %(K) of height 3 remains wide open.
The next result, from [9], is a modest beginning.

276

GEORGE PHILLIP BARKER

THEOREM2.A.25. Let K be an indecomposahle

selfdual cone for which
YT(K) is modular. Let h(K) denote the height of $T(K). Zf dim K =4, then K
is strictly convex.
or h(K)

=3.

Zf dim K ~5,

then either h( K)=2

In the latter case every maximal face

(so K is strictly convex)

is of dimension

3 and is

strictly convex.

The proof is a tedious argument by cases. We close this section with the
conjecture that the case dim K = 5, h(K) =3 is not possible.
B.

The lheory for lI( K)

A given closed full cone K completely determines the cones K and Il( K).
In particular, from a knowledge of the extreme rays of K one should be able
to determine the extremals of K and of II( K ). Actually doing this seems to
be quite difficult. Work in this direction was initiated by Loewy and
Schneider [46].
NOTATION 2.B.l.

Let K be a cone. The set of all extremals of K is

denoted by Ext K.
Thus the set of all extremals of Il( K) is denoted by Ext n(K).
THEOREM 2.B.2.

Let K be a full closed

cone in V. The following

are

equivalent:

(1) K is indecomposable;
(2) if A is nonsingular and A(Ext K) CExt K, then A EExt H(K);
(3) if A is nonsingular
(4) ZEExt H(K).

and A K = K, then A text

II( K );

Before continuing with the discussion of Ext lX( K) let us give another
interpretation of n( K ). If we represent the action of V on V by (f, x ) and
define an inner product in Hom( V ) by (A, B ) = tr ATB, then we may
identify V@ V with Horn (V), where x@ f( y)= (f, y )x. If K 8 K denotes
the set of all bilinear functionals B such that B( x, f ) 2 0 Vr E K, VIE K , then
K 60 K can be identified with Il( K ). There is the obvious extension to tensor
products of cones K, and K, in vector spaces V, and V,. Here of course if x is
a column vector and f is a row vector, we may identify x@f with the matrix
product xf. Berman and Gaiha [20] noted that n( K ) is the closure of the
convex hull of { f@ x: f E K, x E K }. Tam [67] strengthened this considerably.
THEOREM 2.B.3. II( K ) is equal to the set of all nonnegative
combinations of the form zyr with y E K, -E K.

linear

THEORY OF CONES

277

Now let A( K)=II(K)Cn(K).

Thus A(K) is the set of all nonnegative
linear combinations of rank-one matrices yzT with y EK, zEK. Barker and
Loewy [13] established the following theorem.
THEOREM 2.B.4.
lent:

Let K be a closed full cone. The following

are equiva-

(1) K is indecomposable,
(2) K is indecomposable,
(3) If( K ) is indecomposable,
(4) A(K) is inclecomposable.
Let xEExt K, fEExt K'. Then x@fEExtII(K)
and f@zrxExtTI(K).
Following the arguments of Haynsworth, Fielder, and Ptak [40], who presented this for polyhedral cones, one can show that every rank-one extremal is
of this form. When are these all?
To phrase this question in terms of tensor products, consider the following
slightly more general situation. Let K, and K, be cones in V, and V,
respectively. As above let K,@KK, denote the set of nonnegative bilinear
functionals on Vi@ VL. That is,
K@K,={B:

B(f,g)>OVfEK;,gEK;}.

Now define
K,@pK,=

{2xi@yi:

xi EK,,

yi EKE},

where all sums are finite. Clearly K,C3 pK, c K,@ K,. Barker [6] showed that
if one of K, and K, is lattice ordered, then equality holds. Further it was
conjectured in that paper that the converse is true. When K, =K; the
statement can be rephrased as a statement about the self-duality of IX(K).
Tam [67] and Barker and Loewy [13] independently settled this question. We
take the statements from the latter paper.
THEOREM 2.B.5.
simplicial.

Let

1 denote

the identity

matrix.

I EA( K)

iff K is

THEOREM2.B.6. Let K be a cone in R , and let P denote the nonnegative

orthant. Then n( K ) = TI( K ) iff K = QP where Q is an orthogonal matrix.

Thus even when K is polyhedral but not simplicial or when K is the

Lorentz cone, we do not have TI( K) = lI( K ). This implies that in general
fI( K) has many more extremals than just the set of rank-l extremals.

278

GEORGE PHILLIP BARKER

For polyhedral cones Fiedler and Pt&k [30, 311 have given a penetrating
analysis of extreme positive operators. Their main result says- roughly speaking-that
the rank of an extreme positive operator may assume any of the
possible values within certain natural boundaries which they spell out, except.
rank two.
If K is polyhedral with Ext K represented by xi,. . ., xv, set P=[r,
...
x,], the nXr matrix with columns xi. We may associate a similar matrix Q
with K. Bums, Fiedler, and Haynsworth [25] discuss the properties of K and
of lI( K) in terms of these matrices.
The situation for the Lorentz cone is also far from trivial. Let C denote the
Lorentz cone in Iw, n&2. Loewy and Schneider [47] established that Ext C
consists of the rank-one extremals together with the set of maps of C onto
itself. Nor is this situation typical. R. C. OBrien [54] has constructed an
indecomposable K C If3r and a nonsingular A E Ext II( K ) which does not take
Ext K into itself.
Clearly a great deal remains to be done in the analysis of lI( K) for
general K.
The collection of onto maps of K are interesting for another reason. They
form a group called the automorphism group of K, Aut( K). Thus A EAut( K)
iff A->O.
A special class of cones arise in the study of automorphic
functions of several complex variables and also in Jordan algebras.
A selfdual cone K CR is homogeneous iff Aut( K)
DEFINITION2.B.7.
acts transitively on int K.
E. B. Vinberg [74] effected a classification of these cones using a correspondence with compact semisimple Jordan algebras. His main result is that
every indecomposable homogeneous cone is unitarily equivalent to a cone of
one of the following classes:
(I) the cone of n X n real positive semidefinite matrices;
(II) the cone of rr X n positive semidefinite hermitian matrices;
(III) the cone of n X n positive semidefinite matrices over the quaternions;
(IV) the cone of 3 X 3 positive semidefinite matrices with elements from
the Cayley numbers;
(V) the Lorentz cone.
(See also Loewy and Schneider [47] for a discussion of the Lorentz cone.) For
a self-dual cone K in a (not necessarily finite-dimensional) real Hilbert space
N there is a weaker notion of homogeneity. If FZI K, it is easy to check that F LJ
is well defined. Let PF denote the orthogonal projection onto F and set
NF=PF-P+.

THEORY OF CONES

279

Finally let G,( K ) be the real Lie group generated by { NF:

G,( K ) be the Lie group generated by G,( K ).
DEFINITION2.B.8.

A cone K is facially homogeneous

FaK}, and let

iff G,(K) c Aut( K ).

K homogeneous implies K is facially homogeneous, and as is often the

case, the situation in finite-dimensional spaces is nicer. Bellissard, Iochum,
and Lima [18] have established the following result.
Let H be a finitedimensional
THEOREM2.B.9.
cone K is homogeneous iff it is facially

selfdual

real Hilbert space.

Then a

homogeneous.

A discussion of the infinite-dimensional case can be found in a paper by

Bellisard and Iochum 1171.
Since the group Aut( K) contains a copy of the positive reals (the positive
scalar maps), it cannot be compact. Brown [24] considered compact groups.
THEOREM 2.B.10.
matrices

compact

group

of

(elementwise)

nonnegative

is finite.

This result does not extend to arbitrary K. For if we take a Lorentz cone
in Iw3, say

and let
0
COSe

-sin0

sin0 ,
cos 6

then A is a rotation of K around the x1 axis. If 8 is not a rational multiple of 71,

then A generates an infinite compact subgroup of Aut( K ). In fact we have a
more general result.
THEOREM2.B. 11 (Horne [43]).
has a nonzero fixed
Aut( K ).

If A E Aut( K) and if A is irreducible and

point in K, then A belongs

to a compact

subgroup

of

Proof.
If A is irreducible and has a fixed point in K, then its Perron root
p(A) = 1. Since A E Aut( K ), p( A-) = 1 also, as A has (up to scalar multiples)
only one eigenvector in K. Thus all eigenvalues of A have modulus 1 and

280

GEORGE PHILLIP BARKER

linear elementary
divisors (cf. Barker and Schneider
{ A : n = 0, 1,2, . . . } is bounded and hence has compact

[14]).

Thus the group

closure.

This result leads to an extension and alternate proof of the result which
states that a proper Lorentz has an eigenvector in the light cone, that is, the
boundary of a Lorentz cone in R4.
THEOREM 2.B.12.

Suppose K is a full cone in V, where dimV=n

Let A E Aut( K ), and assume

al-C.

that det A >O.

is even.

Then A has an eigenvector

in

Proof.
By dividing A by its spectral radius we may assume that A has a
nonzero fixed point in K. If A has no eigenvector in 8 K, it is irreducible (cf.
Vandergraft
[72]). So by Theorem 2.B.11, A belongs to a compact subgroup
of Aut( K ), whence all eigenvalues are of modulus 1 and are simple roots of
the characteristic
equation of A (Vandergraft
[72] or Barker and Schneider
n
[ 141). Since det A >O, then det A = 1. This cannot happen if n is even.

For polyhedral

cones Browns Theorem

THEOREM 2.B.13.
Aut( K ) is finite.

2.B.10 still holds.

Let K be a polyhedral

cone. A compact

group

G C

The proof of Theorem 2.B.13, which is found in [5], is modeled on

Browns proof of his result, and both rely on the spectral properties
of
nonnegative
operators and some properties of Lie groups. Using Theorem
2.B.13, Horne [43] has obtained
a very nice description
of Aut( K) for
polyhedral K.
THEOREM 2.B.14.

Let K be an indecomposable

polyhedral

cone.

Then

Aut( K ) is isomorphic to the direct product of the group of positive reals (alias
positive scalar maps)

and a maximal finite subgroup.

So far we have considered the structure of Aut( K) C IT( K ). But ll( K)

itself can be considered a semiring; that is, II(K) is closed under addition,
multiplication,
and multiplication
by positive scalars. Left, right, or twosided
ideals of ll( K ) are defined in the obvious way. Thus c C II( K ) is a left ideal
of Il( K ) iff C is closed under addition and AB E I? dA E lI( K ), VB E c. If one
recalls Theorem 2.B.2 it is no surprise that there is a relation between the
indetimposability
of K and the noninvertible
elements of II( K ). Specifically,
Home [42] obtained this result.
THEOREM 2.B.15.
Let K be a full cone in V. Then K is indeconiposable
iff the only maximal right ideal of lII( K) is the set of noninvertible elements.

THEORY OF CONES

281

Proof.
If K is indecomposable,
then by Theorem 2.B.2 ZEE~~ II(K).
Thus the set M of noninvertible
elements of II(K) is closed under addition,
For ZEE~~ II(K) iff some (and hence every) element of Aut( K) is extremal.
Thus the sum of noninvertible
elements is noninvertible.
Therefore M is a
proper (twosided)
ideal of II( K ), and hence the only maximal ideal of any
type. Now suppose K has a nontrivial
decomposition
K = K,@ K,. Define
A,En(K)byA,lspanK,=Z,A,JspanK,=O.DefineAzanalogously.Neither
A, nor A, is invertible, and A,+A,=Z.
Then %(Ai)=Ain(K)
are proper
right ideals, and by a maximality argument each is contained in a maximal
rightideal,say<R(Ai)CL%ifori=1,2.
SinceZ=A,+A2,wehave%,#?Rz,
n
and II( K ) contains at least two distinct maximal ideals.
The cone lI( K) is defined in terms of K, and a vector-space isomorphism
of V clearly induces an isomorphism of lI( K ). To what extent is the converse
trne? Horne [42] considered polyhedral K.
THEOREM2.B.16. Let Vi and V, be finitedimensional
real vector spaces
with full polyhedral cones K, and K, respectively. Suppose T is an isomorphism of the semiring ll( K,) onto H( K,). Then there exists a linear isomorphism

T* : Vi -+ V, mapping

K, onto K, and satisfying

T*(A(x))=T(A)T*(x)

for aZZAETI(K,),

XEK,.

Raphael Loewy in a private communication

has indicated that he has
established this result for nonpolyhedral
cones as well. Tam [69] also extended
Theorem 2.B.16, and we give his results here.
PROPOSITION2.B.17.

ideal,

lI( K ) has a unique

II(K
Furthermore,
II(
by each of its nonzero elements.

namely

generated

no nontrivial

minim&

(nonzero)

two-sided

is a principal

two-sided ideal
Also K is simplicial iff II( K) has

two-sided ideals.

Proof.
It is easy to verify that lI( K) is a twosided ideal. Let $l be a
nonzero twosided ideal of II( K ). Choose a nonzero A EP~. Then there are
yEint K, z Eint K such that (-_)rAy=l.
For any ~EK, zEK we have
yzT=(y@jA(yzr)Cf.
Th us $3 II(K
The next statement
is easy to
prove. If II(K) contains no nontrivial twosided ideal, then II( K )=Il( K ).
The last statement now follows from Theorem 2.B.5, since A( K ) = Il( K). n
THEOREM 2.B.18.
II( K,) are isomorphic

Zf the unique

as semirings,

minimal two-sided ideals of Il( K,) and

then K, and K, are linearly isomorphic.

282

GEORGE PHILLIP BARKER

The ideal II(

also determines whether K is simplicial in another way.

THEOREM2.B.19.

K is simplicial iff TI( K) is a prime two-sided ideal of

n(K).

Although II(K) is not commutative, the definition of prime ideal used

here is this: If PIis a two-sided ideal and if AB E 4 and A @g implies B E 4, then
$ is a prime ideal.
Other natural algebraic questions concerning ideals revolve around principal ideals and maximal ideals. In the full matrix algebra the ascending and
descending chain conditions hold, since right and left ideals are subspaces.
The semiring lI( K) is not so well behaved. Home [42] considered descending
chains, and Tam [69] modified Homes technique for ascending chains. We
can summarize their results as follows.
THEOREM 2.B.20. Let K be an arbitrary cone in a vector space V with
Then n(K)
contains both infinite strictly ascending
and de-

dimVa2.
scending

sequences

of principal

right or left ideals.

For an indecomposable K Theorem 2.B.15 settles the question of maximal

ideals. Horne considered the maximal left and right ideals in the semiring N,,
of elementwise nonnegative matrices. Tam extended these considerations in
the following way. Let K be decomposable, and let K =K,@ . . . &, K, denote
the unique representation of K as a direct sum of indecomposable subcones.
Denote Pi the projection onto span Ki along the sum of the spans of the
remaining faces. Then Pi E II( K ) for all i, and Pi + . . . + P, =I.
THEOREM 2.B.21. With K and Pi as previously
II(K) has exactly r maximal right ideals, namely
$={AEII(K):
and exactly r maximal

described

the semiring,

i=l,...,r;

PieAn(K

left ideals, namely

~,={AEH(K):

Pi en(K)

i=l,...,r.

The proof proceeds along the expected lines except that the proof of closure
under addition employs the fact that the Pi are ertremals in II(K).

Tam also considers twosided

maximal two-sided ideal is one of

ideals in II(A).

He shows that every

THEORY OF CONES

283

Although each -xi, i = 1,. . . , r, is a twosided ideal, not all of these are in
general maximal.
So far we have considered what might be called the arithmetic theory of
ideals. However, ff( K) has another algebraic property, namely the partial
order. The general theory for the faces of II(K) is unsettled at the present.
However, Tam in his thesis and in [69] introduced a simple kind of face.
For F, GET(K)

DEFINITION2.B.22.

set

~I(F,G)=(AEII(K):

AKG}.

Not all faces of fI( K) are of this form, and in particular @( 1 )dff( K) is
not. However, it can be shown that all maximal faces are of this type. Barker
[8] considered the faces fI( F, (0)) and ff( K, F) for a perfect cone (Definition
2.A.15). It is easily seen that fI( F, (0)) is a left ideal and II(K, F) is a right
ideal.
A face of fI( K) which is also a right (left) ideal will
DEFINITION2.B.23.
be called a right (left) facial ideal.
Tam observed that there is a one-to-one correspondence between right
facial ideals of ff( K ) and right ideals in the algebra Hom( V). A similar
remark holds for left and two-sided facial ideals. Thus there are no proper
two-sided facial ideals, although unless K is simplicial, K has proper two-sided
ideals. The ideal structure suggests a classification of the right and left facial
ideals. To verify the details we need some lemmas and notation.
NOTATION2.B.24.

If ScfI(K),

we define ST cII(K)

by

ST= {AT: AES}.

LEMMA2.B.25.

lfF,

G are faces

of K, then

ll( F,G)bI(GD,
In addition

equality

FD).

holds if G is exposed.

Proof
Let AE~(F,G),
~EF,
ZEG~.
Then O=(z,Ay)=(ATz,y),
whence A( G D, c F D. The first statement follows. Assume further that G is
exposed. Taking the transpose of the inclusion, we obtain
lYI( F,G)cIX(G~,F~)?

GEORGE PHILLIP BARKER

284

On the other hand

since G is exposed. Clearly, since F a F Ds, we have

lII(FD, G)cTI(F,G).
Equality

follows.

LEMMA 2.B.26. Let F,, F, E??( K ).

(1) H(K, F,)Cn(K,
F,) iff F,a F2.
(2) lf every face of K is exposed, then
$FD{O))~$F,,{O))

iff

F,aF,.

Proof. We prove (2), and the proof of (1) is similar. In (2) clearly FlaF2
is true. Let
implies ~(F,,{O})>II(F,,{O}).
So assume the latter containment
y be in the relative interior of FzD, and let x E int K. Recall that x8 y E Il( K)
is defined by (x@y)(z)=(y,
z)x. Then ker(xC3y)nK=Fz
from the.discussion following Definition 2.A.9. Thus

whence
F,aker(x@y)nK=Fz.

COROLLARY 2.B.27.
lhen

Let every face of K be exposed, and let F,, F, ET(K).

(1) n(&, (O})+WF,> {O})CWF,r\F,,

(O}),
(2) WK, F,)+WK,
F,)CWK,
F,vF,).
THEOREM 2.B.28. A subset ofll(K)
form II( K, F) for some Fa K.

is a right facial ideal iff it is of the

THEORY OF CONES

285

Proof. Sufficiency is straightforward. Let @(A) be a right facial ideal of

II(K).
We show that @(A)=II(K,
@(Ay))
for any y~int K. Clearly
Q( A)a TI( K, @( Ay)).Choose zEint K. Since a( A) is a right ideal, Ay@zE
(A). But Ay@3z(K\{O}) is contained in the relative interior of @(Ay),
Hence
whence Ay @ z is in the relative interior of II( K, @(Ay)).
lI( K, @(Ay))=@(Ay@~)a@a(A),
and equality follows.
n
THEOREM2.B.29. A subset of H(K) is left facial ideal iff it can be
in the form @(y@z)
with y~int K, ZEK.
If, in addition, the
cot-respondence F -+ F D is surjective, the left facial ideals of II(K) are of the
expressed

form II(F,

(0))

with FaK.

Proof.
To establish the first part note that if @(A) is a left facial ideal of
n(K), then Q(AT)=@(A)r
is a right facial ideal of Il( K). Now modify the
proof of Theorem 2.B.28.
Suppose in addition that F-+ F D is sujective. Then every face of K is
exposed. So from the proof of Theorem 2.B.28 we have

for x lint

K. Hence for F= (a( ATz ))'a K we have

and the proof is finished.

Using these results we may also say something about zero divisors in
n(A).
THEOREM2.B.30. A linear operator A EII( K) is a right (or left) zero
divisor iff A belongs to a proper right (or left) facial ideal.
See Tam [69] for the proof.
We close on the refrain of conditions equivalent to K a simplicial cone.
Recall that if K is perfect then every face is exposed.

THEOREM2.B.31. Let K be a perfect cone. The sum of any two lkft

facial ideals is a left facial ideal when and only when K is simplicial. The
corresponding result also holds for right facial ideals.

286

GEORGE PHILLIP BARKER

Proof. If K is simplicial, we may assume it to be the nonnegative

orthant. It is easy to show that

W5, W)+~(% {O})=~(F,~\F,,

(0)).

Suppose for any F,, F, ET(K) we have that II(F,, {O})+II(F2, (0)) is a left
facial ideal. In particular, if F2 = FF and Fl # {0}, F, #K, then by Theorem
2.B.29 there is a GaK such that

But by Corollary 2.B.27 (I) and Lemma 2.B.26 (l),we have GdF,r\Fro={O},
whence Il(G, {O})=II(K).
Thus there are nonzero Ai EIT(F,, {0}) such that
I =A, + A, and K is decomposable. We check that K = F,@ FID. For any
xEK, x=lx=A~x+A~x.
Further, A<xEFP and A;~EF~, by Lemma 2.B.25
and the fact that FrDD=F. Since it is clear that (span F)n(span FD)= {0},
we have K = F,@ FID.
Now if Fl is any nontrivial face and if x2 EF~ is an extremal, note that

(*>
To verify (*) let ~JEF,~. Then y=axs +x3, where x3 ~@(x,)~,
since K=
@(~,)@@(x,)~.
But then y>x,>O,
so x3 EFF, and thus x3 E@(x,)~/\F~~.
Now start with F, =4)(x1) where x1 is an extremal, and apply (*) n times,
where n = dimV. We obtain K = a( x1) @ * . . $ Ca(x,), and K is simplicial.
The statement for right facial ideals follows from what has been proven
and Lemma 2.B.25.
n

III.

THE CONE OF POSITIVE

SEMIDEFINITE

MATRICES

Let X, be the real space of nXn hermitian matrices, and let ?inbe the
subset of nXn positive semidefinite matrices. The common inner product on
nXn matrices given by (A, B) = tr(B*A) when restricted to x,, makes x,,
into a real inner product space. It is readily checked that ?i,, is a closed full
pointed cone in X,,. It is also well known (cf. Berman 119, p. 551) that ??nis
self-dual. An outstanding problem in this area is a description of IQ??,,) (cf.
de Pillis [57]). In particular, what are the extremals in II(
As we know
from Sec. II.A, we may identify II(??,,) with Yn@q,,, which is not self-dual,
since Yn is not simplicial (Theorem 2.B.6). First we would like to know the

THEORY OF CONES

287

extreme ray of G?nitself. The classification has been part folklore for a number
of years. The version presented here is from Barker and Carlson [lo], where
there are some indications of other references. Let 8, be the set of orthogonal
projections, that is, A ~(3, iff A ~96, and A2 =A. The usual order on
projections is A< B iff range A crange B. It is easily checked that this order
coincides with the partial order 0, inherits as a subset of Tn. 8, is a lattice if
we define AvB and AAB to be the hermitian projections onto, respectively,
range A + range B and range A n range B. Thus e, is isomorphic with the
lattice of all subspaces Q:n. It is also isomorphic to the face lattice 9(qn) of 9,,.
THEOREM
preserving

3.1. The map from Co,to 3( ??,,) given by A + @(A ) is an order

lattice isonwrphism.

We now consider the linear transformations on X,. There are two

representations of the tensor product commonly used, and we must take some
care in distinguishing them. For A, B E %, we have
(A@nB)(C)=tr(BC)A

(thedyadproduct),

(A@,

(the Kronecker product).

B)(C)=ACC

The dyad product is the one which corresponds to the tensor products
used in Sec. II. Thus in our present notation we have IT( 9) = qn@n Tn. We
also know from Theorem 2.B.3 that

Therefore, the extremals of ll(??,,) can be represented as operators of the

form P@, Q where P and Q are rank one orthoprojectors. The extremal
structure of ll( Tn) remains open.
The investigation of I&?) has led to a study of the class of completely
positive operators. A linear operator T on M, is completely positive iff for all
m the map on the set of mXm matrices whose entries are n X n matrices
given by [ Aii] --$ [T( Aii)] preserves the positive semidefinite matrices of
M,@,
M,. Let e,, denote the set of restrictions of completely positive maps to
t7i,. In a forthcoming paper with Richard Hill and Ray Haertel using both
cone theoretic methods and techniques developed by Hill and Poluikis [41]
we are able to describe e,,.
PROPOSITION 3.2. n(??,>)>en >n(??,,)',where for n>2
are strict.

the inclusions

GEORGE

288

PHILLIP BARKER

In fact Choi [26] shows that lI(??,,)=(&

for n=2 and briefly discusses
why the result does not extend. Since JQ is not simplicial, then e, strictly
contains II( Tz).

THEOREM

3.3. C?,,
is isometrically
matrices.

isomorphic

with

9,,?, the cone

of

n2 X n2 positive semidefinite

Theorem (3.3) indicates why in many respects it is e,, and not II(??,,)
which is the natural notation of a positive operator for maps which preserve
the nXn hermitian matrices.
REFERENCES
6. P. Barker, Monotone norms, Numer. Math. 18:321-326 (1972).
G. P. Barker, On matrices having an invariant cone, Czechoslouak Math. J.
22:49-68 (1972).
G. P. Barker, Matricial norms over cones, Linear Algebra and AppZ. 6:175- 182
(1973).
4 G. P. Barker, The lattice of faces of a finite dimensional cone, Linear Algebra and
Appl. 7:71-82 (1973).
5 G. P. Barker, Stochastic matrices over cones, Linear and Multilinear Algebra
1:279-287 (1974).
6 G. P. Barker, Monotone norms and tensor products, Linear and Multilinear
Algebra 4, No. 3. (1976).
7 G. P. Barker, Faces and duality in convex cones, Linear and M&linear Algebra
6:161-169 (1978).
8 G. P. Barker, Perfect cones, Linear Algebra and AppZ. 22:211-221 (1978).
9 G. P. Barker, Modular face lattices: low dimensional cases, Rocky Mountain J.
Math., to appear.
10 G. P. Barker and D. Carlson, Cones of diagonally dominant matrices, Pacific J.
Math. 57:15-32 (1975).
11 G. P. Barker and D. Carlson, Generalizations of topheavy cones, Linear and
Multilinear Algebra 8:219-230 (1979/80).
12 G. P. Barker and J. Foran, Self-dual cones in Euclidean spaces, Linear Algebra
and Appl. 13: 147- 155 (1976).
13 G. P. Barker and R. Loewy, The structure of cones of matrices, Linear Algebra
and Appl. 12:87-94 (1975).
14 G. P. Barker and H. Schneider, Algebraic Perron-Frobenius theory, Linear
Algebra and AppZ. 11:219-233 (1975).
15 G. P. Barker and R. E. L. Turner, Some observations on the spectra of cone
preserving maps, Linear Algebra and Appl. 6:149- 15.3 (1973).
16 F. L. Bauer, Lecture Notes on Positivity and Norms; Part I: Ordering and
Positiuity, Technische Universitgt Miinchen, (1974).
17 J. Bellissard and B. Iochum, Homogeneous self dual cones versus Jordan algebras,
Ann. ht. Fourier 28:104-144 (1978).
18 J. Bellissard, B. Iochum, and R. Lima, Homogeneous and facially homogeneous
self-dual cones, Linear Algebra and Appl. 19:1- 16 (1978).

THEORY
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36

37

38
39
40
41

OF CONES

289

A. Berman, Cones, Matrices, and Mathematical Programming, Lecture Notes in

Economics and Mathematical Systems 79, Springer, New York, (1973).
A. Berman and P. Gaiha, A generalization
of irreducible monotonicity,
Linear
AZgebru nnd A$.
5:29-38 (1972).
A. Berman and R. J. Plemmons, Nonnegative Matrices in the Mathematical
Sciences, Academic Press, New York, 1979, Chapters 1, 5, and notes thereof.
G. Birkhoff, Linear transformations
with invariant cones, Amer. Math. Monthly
72~274-276 (1967).
G. Birkhoff, Lattice Theory, 3rd ed., Amer. Math. SOC. Colloq. Publ. 25, Amer.
Math. Sot., Providence, R. I., (1967).
D. R. Brown, On clans of non-negative
matrices, Proc. Amer. Math. Sot.
15:671-674 (1964).
F. Burns, M. Fiedler, and E. Haynsworth,
Polyhedral cones and positive operators, Linear Algebra and Appl. 8x547-559 (1974).
M.-D. Choi, Completely
positive linear maps on complex matrices, Linear
Algebra and Appl. 10:285-290 (1975).
P. Crawley and R. P. Dilworth, Algebraic Theory of Lattices, Prentice-Hall,
Englewood Cliffs, N.J. (1973).
L. Elsner, Quasimontonie und Ungleichungen
in halbgeordneten
Raumen, Linear
Algebra and Appl. 8:249-261 (1974).
M. Fiedler and E. Haynsworth,
Cones which are topheavy with respect to a
norm, Linear and Multilinear Algebra 1203-212
(1973).
M. Fiedler and V. Pt&
Diagonals of convex sets, Czechoslouak Math. J.
28(103):25-44
(1978).
M. Fiedler and V. Pt& The rank of extreme positive operators on polyhedral
cones, Czechoslovuk Math. J. 28(103):45-55 (1978).
G. Frobenius, iiber Matrizen aus positiven Elementen, S.-B. Preuss. Akad. Wiss.
Berlin, 1908, pp. 471-476.
G. Frobenius, liner Matrizen aus positiven Elementen,
II, S.-B. Preuss. Akad.
Wiss. Berlin, 1909, pp. 514-518.
G. Frobenius, iiber Matrizen aus nicht negative Elementen, S.-B. Preuss. Akud.
Wiss. Berlin, 1912, pp. 456-477.
M. Gerstenhaber,
Theory of convex polyhedral cones, in Activities Analysis of
Production and Allocation (T.C. Koopmans, Ed.), Wiley, New York, (1951).
I. M. Glazman and Ju. I. Ljubic, Finite-Dimensional Linear Analysis, (G. P.
Barker and G. Kuerti, Transl. and Ed.), MIT Press, Cambridge,
Mass., 1974,
Chapters 7, 8.
A. J. Goldman and A. W. Tucker, Polyhedral Convex Cones in Linear inequalities
and Related Systems (H. W. Kuhn and A. W. Tucker, Eds.), Princeton U. P.,
Princeton, N.J., (1956).
B. Griinbaum, Convex Polytopes, Wiley, New York, (1967).
E. Haynsworth,
A note on cones of positive operators, Abstract, Notices Amer.
Math. Sot., 1969, pp. 667-135.
E. Haynsworth, M. Fiedler, and V. Ptak, Extreme operators on polyhedral cones,
Linear Algebra and Appl. 13: 163- 172 (1976).
R. D. Hill and J. A. Poluikis, Completely positive and Hermitian preserving linear
transformations,
Linear Algebra and A&., to appear.

290
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59

60
61
62
63
64

GEORGE

PHILLIP

BARKER

J. G. Home, On ideals of cone preserving maps, Linear Algebra and AppZ.

21:95-110 (1978).
J. G. Home, On the automorphism
group of a cone, Linear Algebra and Appl.
21:111-122 (1978).
L. Kantorovitch, Snr les espases semiordonnes lineaires et lenrs applicationes a la
theorie des operationes lineaires, C.R. deZAc. Sci. LJRSS 4:13-16 (1935).
M. G. Krein and M. A. Rutman, Linear operators leaving invariant a cone in
Banach space, Amer. Math. Sot. Trunsl. Ser. 1 10:199-325 (1950).
R. Loewy and H. Schneider, Indecomposable
cones, Linear Algebu and AppZ.
11235-245
(1975).
R. Loewy and H. Schneider, Positive operators on the ndimensional
ice-cream
cone, Z. Math. Anal. AppZ. 49:375-392 (1975).
P. McMullen and G. C. Shephard, Conoex Polytopes and the Upper Bound
conjecture, Cambridge U.P., London, (1971).
I. Marek, We-positive operators and some of their applications,
SIAM J. AppZ.
Math. 15:484-494 (1967).
I. Marek, Spektrale Eigenschaften
der K-positiven Operatoren
und Einschliessnngsatze fiir den Spektralradins,
Czechoslovuk Math. J. 16:493-517 (1966).
I. Marek, Frobenius
theory of positive operators: comparison
theorems and
applications, SIAM J. AppZ. Math. 19:607-628 (1970).
I. Marek, A note on K-stochastic
operators, Cusopis P&t. Mat. 96~239-244
(1971).
W. Meixner, Generalized norms, dissertation, Technische Universitat Miinchen,
(1977).
R. C. OBrien, On extreme matrices and extreme vectors of cones in R n, Linear
Algebra and AppZ. 12:77-79 (1975).
J. M. Ortega and W. C. Rheinboldt, Iterative Solutions of Nonlinear Equations in
Several Variables, Academic, New York, (1970).
0. Perron, Znr Theorie der Matrices, Math. Ann. 63:248-263 (1907).
J. de PiIlis, Linear transformations
which preserve hermitian and positive semidefinite operators, Pacific J. Math. 23:129- 137 (1967).
N. J. Pullman, A geometric approach to the theory of nonnegative
matrices,
Linear Algebra and Appl. 4:297-312 (1971).
W. C. Reihnboldt
and J. S. Vandergraft,
A simple approach to the PerronFrobenius
theory for positive operators on general partially ordered finitedimensional linear spaces, Math. Camp. 27:139-145 (1973).
H. Schneider, Positive operators and an inertia theorem, Numer. Math. 7:11-17
(1965).
H. Schneider and M. Vidyasagar, Cross-positive matrices, SIAM I. Numer. Math.
7508-519
(1970).
J. Schriider, M-matrices and generalizations
using an operator theory approach,
SIAM Rev. 20:213-244 (1978).
J. Stoer and C. Witzgall, Convexity and Optimization in Finite Dimensions I,
Springer, (1970).
B.-S. Tam, A note on cross-positive matrices, Linear Algebra and A&. 12:7-9
(1975).

THEORY

OF CONES

291

65

B.-S. Tam, A note of polyhedral

(1976).

66

B.-S. Tam, Some results on cross-positive matrices, Linear Algebra and Appl.
15:173- 176 (1976).
B.-S. Tam, Some results on polyhedral cones and simplicial cones, Linear and
Mu&linear Algebra 4:281-284 (1977).
B.-S. Tam, On indecomposable
subcones, Linear Algebra and A&. 22:33-40
(1978).
B.-S. Tam, On the semiring of cone preserving maps, Linear Algebra and Appl.
35:79- 108 (1981).
U. Trottenberg
and G. Winter, On faces and W-irreducible operators in finite
dimensional linear spaces, Linear Algebra and Appl. 33:87- 110 (1980).
H. Uzawa, A theorem on convex polyhedral cones, in Studies in Linear and
Nonlinear Progrumming (K. J. Arrow, L. Hurwicz, and H. Uzawa, Eds.), Stanford
U. P., Stanford, Calif., (1958).
J. S. Vandergraft,
Spectral properties of matrices which have invariant cones,
SIAM J. Appl. Math. 16:1208-1222 (1968).
J. S. Vandergraft, A note on irreducibility for linear operators on partially ordered
finite dimensional vector spaces, Linear Algebra and AppE. 13:139- 146 (1976).
E. B. Vinberg, The theory of convex homogeneous cones, Trans. Moscow Math.
Sot. 12:340-403 (1963).
H. Wielandt, Unzerlebare nicht-negative
Matrizen, Math. Zeitsch. 52:642-648
(1956).

67
68
69
70
71

72
73
74
75

cones, J. Austral.

Receioed 1 April 1980; revised I7 November 1980

Math. Sot. Ser. A 22:45&461